The beat error is a hot item on many watch-related forums.
If you look it up, you might find something like: “beat error is an asymmetrical oscillation of the balance wheel.”
In this article, I’ll explain what it is and why it’s important.
What exactly is the beat error?
The beat error is the difference between the clockwise swing and the counterclockwise swing of the balance. In perfect conditions, the beat error is 0.
As you know, the balance wheel swings clockwise and counterclockwise. Each one of those swings, in either direction, is called a “beat.” If the watch runs perfectly, the clockwise swing will be the same as the counterclockwise swing, and both swings will take the same amount of time.
However, perfect conditions don’t exist in the real world.
In reality, the clockwise swing, for example, will take a bit longer than the counterclockwise swing. The watch is slightly “out of beat.”
In the example above, the beat error could be 1.2 ms (it’s measured in milliseconds). That means that the clockwise swing takes 1.2 ms longer than the counterclockwise swing.
Do you have a watch that doesn’t start ticking on its own but needs some “help”? That might be due to a high beat error.
What causes it?
The balance wheel swings to the left and the right. Halfway, the impulse pin of the balance enters and exits the fork of the lever.
The lever is placed between two banking pins.
The zero position of the balance is the position in rest without any power in the mainspring. The location of the impulse pin is most important because that’s the middle point between a clockwise swing and a counterclockwise swing.
The dead center of the banking pins is the ideal resting place for the impulse pin in the zero position. The beat error would be 0.
However, it’s nearly impossible to achieve this so the impulse pin will likely be a little to the left or right of this point. That causes a beat error because one direction of the swing is slightly longer than the other one.
Remember that we’re talking milliseconds here, so the differences are minimal.
How do you correct it?
Mobile stud carrier
If you’re lucky, your watch has a mobile stud carrier. With it, you can quickly reposition the complete balance and therefore the position of the impulse pin in relation to the banking pins.
Place your watch on a timing machine so you can manipulate the mobile stud carrier and monitor the beat error at the same time.
Fixed stud carrier
If the movement doesn’t have a mobile stud carrier, you’ll need to move the collet of the hairspring on the balance staff. That will change the position of the impulse pin in the zero position as well.
Always remove the balance and place it on a balance tack if you want to move the collet. You can use a tiny screwdriver, or you can use a special hairspring collet tool like the Bergeon 30017.
As you can imagine, this is a very fiddly job so you should only consider this when it’s necessary. Otherwise, it’s not worth the risk of damaging the hairspring, especially if you know that spare parts are impossible to find.
What’s a good beat error?
To be clear, the beat error on its own means very little. You should always consider the amplitude and the rate as well.
A watch with a high beat error can still keep accurate time if it has a strong amplitude and a consistent daily rate. However, it’ll be more vulnerable to problems with timekeeping when the amplitude is low or during sudden changes in the position of the watch.
If your movement has a mobile stud carrier, you can easily adjust the beat error to 0.2 ms or 0.3 ms. If it has a fixed stud carrier, a beat error of 0.8 ms and below is fine.
Have you ever corrected the beat error? If you have any more tips, please leave them in the comments below.
22 thoughts on “Watch Talk: What’s the Beat Error of a Watch?”
please ignore the previous issue, I was misinterpreting the design of the regulator arm. Now I have to figure out where the resistance is coming from on the balance assembly. The hairspring is level but I am not sure the part that has the stud is clearing the coils. I ordered this HS from a guy on ebay who claims to have loads of factory Hampden parts specific to my sz 16 Hampden William McKinley. I am putting this watch away for now. As I said it has a host of issues stemming from impact. IT will not run face down only FU and PU, PD . The balance cock hole jewel is slightly off center and I know that is an issue. The other balance jewel may have issues but I didn’t see any on examination. The watch was cleaned, oiled, reassembled and is now in its current state. I don’t have the equipment to correct endshake. The bezel set balance jewel screws are tight and I can’t budge them. It will be put away for another day. Sorry to have taken your time.
It’s not a problem at all. To begin, I would suggest finding a movement that keeps good time, then work your way up. It’s nearly impossible to tackle a watch with so many issues at the same time because you don’t know where to start. When so much is going on, you don’t know if you’ve made any improvements since the watch still doesn’t run correctly.
I should note that all these stats for the pocketwatch are coming from my timegrapher.
I have a pocketwatch with a very high beat error 8 ms. IT has a host of problems. I don’t intend to ever use the watch except as a learning platform. IT loses 5 min a day. The amplitude is over 300 fully wound and comes down as the day wears on.
But here is my question – why when the regulator is fully in the slow position, and it is losing 5 min a day, why does it go even slower when the regulator is fully in the fast position?
I took some care in putting a new hairspring on and aligning the stud of the HS to the roller jewel.
What is going on here, how can it run slower as I reduce the length of the HS via the regulator?
My watch has a high beat error (1.4-1.8) but it is accurate enough (between +3/+5 seconds per day). Is this normal?
No, it’s not normal but it’s common. It means that the watch probably takes a bit longer to start running when you wind it and it’s more vulnerable to position changes.
I have a new Powermatic 80 movement that won’t start no matter how many winds. It only starts when I give it a gentle shake or pull the crown out then push it back. It’s very accurate and the beat error is only about 0.3ms when sufficiently wound. Any idea why this happens?
Sorry, there’s no way to tell. It would be best if you took it to a local watchmaker.
A very interesting and informative article and also, question responses.
Hi Melvin, two questions, thank you for addressing this. The procedures of regulating/adjusting a movement defies my logic. One adjusts the watch in different positions, so if you say, have +24 lying flat dial up, but vertical, crown up -12, will not adjusting on of those positions maladjust the other position? Can a watch maker adjust each position independent of the other? Second, my Tudor GMT crown down is -1 with beat error of .4, and crown up it is -2 with .1 beat error. I am left handed, meaning that I like to wear my watch on my right hand, but wearing on my left hand (crown up) will give me lower beat error. Which is better with the stats I have above?
if beat error can only be adjusted with a timing machine, how did watchmakers regulate this before they had timing machines? was it just trial and error or doing it by eye or is there some kind of antique timing machine?
Yes, you can do it by eye. The balance spring should be fully relaxed in the neutral position and the impulse pin should rest perfectly between the banking pins. However, we’re talking milliseconds here, so this is a crude method.
Thank you for your input. I serviced this movement. All the gears in the drivetrain were clean and straight. I’m pretty sure I oiled everything correctly. I oiled just the flat spots on the pallet fork after I cleaned it. I cleaned the balance assy and oiled the jewel in the balance cock. It runs so good under full power that I’m left scratching my head about the change in beat error. Any suggestion on where to look would be greatly appreciated.
Sorry, I’m at a loss as well. I’ve never seen a sudden change in beat error like that. I’d remove the balance and clean the balance pivots with jewelers rouge or something. Then, clean it (the cap jewels as well) and lubricate again. Did you check if the banking pins are ok? Are the regulator pins ok? Is the hairspring flat and not touching anything? Are all the windings of the hairspring concentric without touching each other?
Very good article. I also have a question. Does the beat error increase as the power of the mainspring decreases? I have a watch that has used most of it’s reserve sitting on the bench. The amplitude and speed are good, 240 and +7. The beat error has gone from 0.2 to 2.5. Thanks
No, it shouldn’t increase because the beat error refers to the position of the impulse pin in rest and the difference in swings. That has nothing to do with the power reserve, especially since the rate and the amplitude are still good. Something else must have caused the beat error to increase.
If a brand new watch has a beat error of 1.1, should it be sent back?
A beat error of 1.1 is too much. I don’t know if the best action would be to send it back though. An independent watchmaker with a timing machine could easily fix it.
I agree with Melvin, very helpful. One question, a close friend of mine received a watch but it was running slow (about 5-10 minutes per day). He put it on a timegrapher and found that the beat error was high. So he tried to fix the beat error first before adjusting the rate. Unfortunately, while adjusting the Mobile Stud Carrier (MSC), the watch stopped running. He wanted to know if its possible to adjust the MSC too much such that the watch stops completely.
Hi, yes that’s possible. A beat error is measured in m/s so it only needs tiny adjustments. It’s a good idea to look up pictures of the movement online to see the general position of the stud. From there, only small corrections are needed. By the way, a rate of -5 to -10 minutes a day is huge.
This was very clear, concise, and helpful, thanks!
Thanks, Joshua! I’m glad it was helpful.
Comments are closed.